Factors associated with parental burden among parents of children with food allergies in China: a cross-sectional study.

OBJECTIVE: This study aimed to investigate the factors associated with parental burden among parents of children with food allergies (FA) in China.
DESIGN: Cross-sectional study.
SETTING: The participants were recruited through the internet and one child health centre of a tertiary hospital in Changsha, China. PARTICIPANTS: A total of 346 parents of children with FA were recruited using a convenience sampling method. After excluding the invalid questionnaires, 330 questionnaires were included for the analysis. OUTCOME MEASURES: The data of general and disease-specific information of children with FA and their parents, as well as parental burden, social support and coping style of parents, were collected using a self-developed questionnaire and standard measurement tools with good psychometric properties. Univariate analysis and multivariate stepwise regression analysis were used to analyse the data.
RESULTS: Financial affordability regarding their children's FA and parents' social support were key protective factors against parental burden (β=-0.224, p<0.001 and β=-0.226, p<0.001, respectively). This study also found that parents' working conditions, the number of highly suspected food allergens, the number of children's FA-affected systems and whether children took amino acid milk powder were the factors associated with parental burden.
CONCLUSIONS: Health professionals should pay more attention to parents who are not at work and whose children have more highly suspected food allergens and more FA-affected systems because these parents often have a higher parental burden. Furthermore, it is important to improve financial support for children's FA and social support for parents of children with FA to decrease their parental burden. © Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

We simultaneously included parents’ demographic characteristics, social support and coping style in this study to explore the factors influencing parental burden among parents of children with food allergies (FA).

The participants were recruited through a convenience sampling method in two ways: through the internet and from a child health centre in a tertiary hospital in Changsha, China.

The coping style measurement employed in the study consider only two coping styles (ie, active and passive), and more comprehensive tools should be considered to accurately identify additional types of coping styles for fuller understanding of parental burden.

A longitudinal study with a larger sample size is needed to explore the influencing factors of parental burden among parents of children with FA.

Background

Food allergy (FA) is a common allergic disease in childhood; it is a hypersensitive reaction of the immune system to proteins in food. It can affect the skin and gastrointestinal, respiratory and cardiovascular systems with varying degrees of severity.1 Children with FA often have multiple allergic comorbidities, such as asthma, eczema and hay fever.2 FA affects approximately 5% of global children aged 0–5 years diagnosed by open oral food challenge (OFC) testing,3 and it has become one of the fastest-growing public health problems with epidemiological characteristics, especially in developed countries.4–6 The prevalence of FA in children has also been increasing at an alarming rate recently in China.7 8 FA requires strict avoidance of food allergens. Accidental exposure to food allergens can endanger children’s health if effective avoidance measures are not taken.9 Avoiding these allergens is not simple, as many products contain ‘hidden ingredients’ or traces of food allergens, such as nuts.10

For younger children with FA, the disease burden not only falls on the child but also on the parents.11–13 Parents of children with FA may be particularly concerned about their children’s allergies and take on additional burdens in their daily lives, as they often play an important role in managing their children’s allergies. Specifically, studies have shown that families with children with FA spend more time and energy on grocery shopping because avoiding allergens requires caregivers to double-check food labels for allergens. Preparing food also increases extra work for families with children with FA, as parents need to prepare separate meals for allergic children and prevent cross-contamination.13 Parents also report its negative effects on daily activities and socialisation.11 12 14 There is growing evidence that families with children with FA have higher financing costs than those whose children are without FA.15 In addition to practical and financial needs, research suggests that FA is also associated with negative psychological consequences for parents, including greater stress and worry.11 FA does hurt parental health-related quality of life.16 This may lead to a decrease in parental caring capacity and affect the development of children.

Evidence in the literature shows that the younger the child, the greater the burden of care on the parents.11 Furthermore, the severity of FA, the presence of concomitant allergic diseases, the number of food allergens and children’s allergy history also influence parental burden.17–19 Regarding parent-related factors, the gender of the parents (father/mother) influenced the caregiving burden. Four studies compared the caregiving burden between parents and found that mothers had a higher caregiving burden than fathers in at least one area.18–21 Nevertheless, existing studies on the factors influencing parental burden among parents of children with FA have been limited to factors related to children with FA, neglecting the influence of other factors, such as demographic characteristics (eg, age, gender and education degree) and other changeable variables (eg, social support and coping style) of parents. Social support and coping style are considered two changeable variables that may influence the parental burden on parents of children with FA. Social support acts as an available external resource against stress.22 Good social support from family, society and others can reduce an individual’s stress and thus may lower caregivers’ burden.23 In addition, coping style refers to an individual’s efforts to change cognition and behaviour to meet specific internal and external environmental requirements beyond their existing capacity and resources.24 Caregivers with an active coping style could better cope with their problems and thus may reduce the burden of caring for children with FA.25 However, there is a lack of research exploring the relationship between social support/coping style and parental burden among the parents of children with FA.

In addition, the literature on this topic in China is limited. Only one study focused on parental quality of life in families with children with FA in Hong Kong, a developed city in China. However, this article emphasised the validation and discriminative ability of the traditional Chinese version of the Food Allergy Quality of Life-Parental Burden (TC-FAQL-PB) questionnaire in Hong Kong. It only used univariate analysis to identify the influencing factors, so the results were yet to be verified. Furthermore, because of the culture, society and economic differences between Hong Kong and mainland China, the results may not be representative, and the associated factors of parental burden in the context of mainland China need to be further explored. Recently, the authors of this study developed and validated the simplified Chinese version of the Food Allergy Quality of Life-Parental Burden (SC-FAQL-PB) questionnaire, which has a wider range of applications than TC-FAQL-PB in China. It has been proven to have good psychometric properties and can be used as an effective measure in further studies.

Therefore, this study aims to explore the factors associated with parental burden among parents of children with FA in China, considering general information about children and parents and the social support and coping styles of parents.

Methods

Study design

A cross-sectional study was conducted to explore the factors influencing parental burden among parents of children with FA in China. This study was reported following the Strengthening the Reporting of Observational Studies in Epidemiology checklist (online supplemental file S1).

Setting and participants

FA tends to commonly occur in early childhood, first occurring before the age of 5 years,26 and the parents of the children aged 0–5 are largely responsible for managing their child’s FA and thus often experience a high parental burden. Therefore, we identified parents of a child aged 0–5 years with FA as our study population. A total of 346 parents of children aged 0–5 years diagnosed as having FA were recruited through convenience sampling of the internet and a child health centre in a tertiary hospital in Changsha, China. There was no significant difference between participants recruited via these two methods. Inclusion criteria were as follows: (1) a parent aged ≥18 years who (2) had at least one child aged ≤5 years who were clinically diagnosed as having FA by a clinician. OFC, skin prick test, spot patch test, serum IgE detection or dietary avoidance can be used in the diagnosis, (3) was the main caregiver of children and (4) could understand and communicate normally. Exclusion criteria were as follows: (1) a parent whose child was complicated with malignant diseases or other organic lesions and (2) a parent whose child had a mental illness or cognitive impairment.

The sample size was calculated using an independent variable-to-subject ratio reported by Kendall et al.27 He recommended at least a 1:10 ratio for a cross-sectional investigation. Moreover, the ‘rules of thumb’ for determining the sample size for regression equations using six or more predictors stated that an absolute minimum of 10 participants per predictor variable was appropriate.28 The independent variables in this study included general information, social support and coping style. Considering the 28 independent variables and a 15% invalid rate, a minimum of 330 cases was required. Thus, a sample size of 346 in this study met this requirement.

Recruitment and data collection

Before completing the questionnaire, the participants were informed of the research-related information, including the research’s objectives and investigation process, anonymity and voluntary participation. The questionnaires were completed on the Internet using Wenjuanxing, an online questionnaire collection platform in mainland China. Wenjuanxing ensured that respondents completed the entire questionnaire to avoid missing data. It also avoided the problems of manual data entry and improved the efficiency and accuracy of data collection and screening. Those who had difficulty filling out the electronic questionnaire could contact researchers to guide them to finish the e-questionnaire or ask for a paper questionnaire. The researchers’ contact information was provided to the participants at the time of recruitment. Before the formal survey, we selected 10 subjects who met the study inclusion criteria for the presurvey. The presurvey ensured that the general information items and questionnaire instructions developed by our research team were logical and easy to understand. Data collection began in December 2021 and lasted for 3 months.

Measures

General information questionnaire

A general information questionnaire consisted of items to gather details about the children’s and parents’ demographic information and other disease-specific characteristics of children with FA. The children’s information included months of age, gender, growth standards (using Z-scores), the number of identified food allergens, the number of highly suspected food allergens, the number of children’s FA-affected systems, the number of FA-related symptoms, the time of the first diagnosis of FA, the frequency of adverse reactions, whether they were going to kindergarten, whether they were taking breast milk, whether they had an intake of special formula milk powder, whether they had an egg allergy, whether they had a milk allergy, whether they had experienced severe anaphylaxis,29 30 whether they had a history of antiallergy drug use (referring only referred to antihistamine use) and whether they had visited the emergency room. Parents’ information included gender, age, parental history of self-reported allergies, education level, working conditions, financial affordability regarding their children’s FA, knowledge about FA (self-reported), whether they were obtaining learning knowledge about FA from health professionals and whether they had a medical knowledge background.

SC-FAQL-PB questionnaire

The FAQL-PB questionnaire was developed by Cohen et al.31 The original English version has been translated into many languages and validated in the UK, Australia, Thailand, Iran, South Korea, Hong Kong and mainland China. The authors adapted and validated the Chinese version, SC-FAQL-PB. It uses a 7-point Likert scale ranging from 1 (not troubled) to 7 (extremely troubled). There are 17 items covering family and social activities, meal preparation, health issues and emotional issues, measuring two dimensions of limitations in life (three items) and emotional distress (14 items). Items for the measurements of parental burden were negatively worded (eg, ‘How much would your choice of vacation be limited by your child’s food allergy?’). A higher SC-FAQL-PB score indicates a higher parental burden. The confirmatory factor analysis results confirmed the two-factor model proposed by Leung et al as the most reasonable fit for explaining the factor construct of the SC-FAQL-PB.30 The internal coefficient of consistency was 0.946, with a test–retest reliability of 0.926. The Cronbach’s alpha of the scale in our sample was 0.958.

Social Support Rating Scale

The social support level was assessed with the Social Support Rating Scale (SSRS), which was developed by Xiao.32 It includes 10 items measuring three dimensions: subjective support (three items), objective support (four items) and support usage (three items). Objective support refers mainly to visible practical support, such as direct material assistance and social networks. Subjective support refers to an individual’s emotional feelings and satisfaction with being supported, respected and understood in society. Support usage refers to the extent to which individuals utilise social support.33 A higher total score indicated a higher social support level. This total support score was classified into three categories: low (≤22), moderate (23–44) and high (≥45) levels.34 The SSRS has been widely used in Chinese domestic contexts.35–37 The cumulative contribution rate of the three common factors was 54.26%.38 The internal consistency of the SSRS in this study was acceptable. The Cronbach’s alpha of the scale in our sample was 0.778.

Simplified Coping Style Questionnaire

Coping style was accessed with the Simplified Coping Style Questionnaire (SCSQ), which is a self-report survey that includes two subscales: active coping (12 items) and passive coping (eight items).39 All items were rated using a 4-point Likert scoring method, ranging from 0 (never) to 3 (always). The scores of each subscale item were summed up to generate a subscale score. Dai proposed a formula for the coping style tendency—coping style tendency=active coping criterion score (Za score)−passive coping criterion score (Zp score)—to determine an individual’s main coping style in stress situations.40 Za/Zp=(actual score for the subscale−mean of the score for the subscale among the national norm sample)/(SD of the score for the subscale among the national norm sample). If the value of the coping style tendency >0, it means the individual mainly adopts an active coping style. The Chinese national norm for the active coping score is 1.78±0.52 and for the passive coping score, it is 1.59±0.66.39 The SCSQ has been widely used to assess caregivers’ coping styles in China, and the Cronbach’s alpha coefficiency of the two subscales ranged from 0.75 to 0.90.41–45 The cumulative contribution rate of the two common factors was 48.5%.46 In the present study, the Cronbach’s alphas of the active and passive coping styles were 0.909 and 0.752, respectively.

Data analysis

The collected data were analysed using the IBM SPSS V.26.0 programme. Descriptive statistics were used to investigate the demographic and disease-specific characteristics of children with FA and their parents, social support, coping style and parental burden among the parents of children with FA. The Kolmogorov-Smirnov test was performed to confirm the normality of the measurement data. Means with SDs (mean±SD) were calculated for normally distributed data, while median and IQRs were calculated for non-normally distributed measures. Frequencies and percentages were calculated for the count data. Between-group differences were tested using t-tests and analysis of variance for normally distributed data to investigate parental burden. Spearman’s correlation analysis was used to analyse the correlations between parental burden, coping style and social support among the parents of children with FA. All reported p values were two-sided, with a significance level of 0.05.

Multivariate stepwise regression analysis was performed to identify the factors influencing parental burden among parents of children with FA, with the total FAQL-PB score as the dependent variable. The variables with a p<0.1 for univariate analysis were all introduced. Some non-normally distributed continuous variables were log-transformed (y=log10 (x+1)) to normalise their distribution before their introduction to multivariate stepwise regression analysis. The variables were screened stepwise based on the criteria of α≤0.05 in and α≥0.1 out model. When the independent variables were unordered categorical variables, they were converted into dummy variables and then included in the regression analysis. The analysis used R2 as the power of a linear model to represent the percentage of the current variation of y explained by the independent variables in the current model.

Patient and public involvement

None.

Results

General characteristics of children with FA and their parents

A total of 346 parents of children with FA were enrolled in this study, with a response rate of 100%. Sixteen questionnaires were removed because the filling time was less than 5 min (six questionnaires) and because the parents were not the primary caregivers of children in their daily lives (10 questionnaires). The valid questionnaire rate was 95.38% (330/346). Table 1 shows the characteristics of the children with FA and the relationships between the characteristics and parental burden among parents. Most children (75.5%, 249/330) had milk allergies and 46.7% (154/330) of the children were allergic to eggs. Only three children had experienced severe anaphylaxis and 21.8% (72/330) had visited the emergency room. As presented in table 2, the male-to-female ratio of parents was 1:16.4, their mean age was 30.41 years and most completed college or had a higher degree (87.9%, 290/330). More than half of the parents (70.0%, 231/330) were at work and most parents (77.9%) reported that the financial costs caused by FA were affordable. More details can be found in tables 1 and 2.

Table 1

General characteristics of children with food allergies and the relationship between the characteristics and parental burden (n=330)

Characteristics		Parental burden
		Mean	SD	t/F/r	P value
Gender, n (%)				0.62	0.54
Male	187 (56.7)	65.19	21.24
Female	143 (43.3)	63.85	18.18
Whether they were going to kindergarten, n (%)				−1.55	0.12
Yes	10 (3.0)	55	19.16
No	320 (97.0)	64.91	19.93
Growth standards, n (%)				−1.02	0.31
Without growth retardation	313 (94.8)	64.35	20.05
With growth retardation	17 (5.2)	69.41	17.99
Whether they were taking in breast milk, n (%)				0.31	0.76
Yes	77 (23.3)	65.22	20.82
No	253 (76.6)	64.42	19.72
Whether they had an intake of special formula milk powder, n (%)				3.39	0.02*
Moderately hydrolysed formula	64 (19.4)	60.69	18.97
Deep hydrolysed formula	106 (32.1)	63.75	20.21
Amino acid milk powder	75 (22.7)	70.67	17.97
Nonea	85 (25.8)	63.29	21.15
Whether they had an egg allergy				2.45	0.02*
Yes	154 (46.7)	67.42	17.57
No	176 (53.3)	62.15	21.28
Whether they had a milk allergy				1.4	0.16
Yes	249 (75.5)	65.91	20.08
No	81 (24.5)	61.91	19.44
The time of the first diagnosis of FA (month of age), n (%)				0.76	0.45
0–5 months	260 (78.8)	65.04	19.68
>5 months	70 (21.2)	63	21.01
The frequency of adverse reactions, n (%)				6.84	＜0.01**
1–5 times	154 (46.7)	62.77	20.3
6–10 times	69 (20.9)	60.07	17.63
>10 times	107 (32.4)	70.18	19.8
Whether they had experienced severe anaphylaxis,b n (%)				0.24	0.81
Yes	3 (0.91)	67.33	26.86
No	327 (99.1)	64.58	19.94
Whether they had a history of antiallergy drug use,c n (%)				1.98	＜0.05*
Yes	140 (42.4)	67.14	19.1
No	190 (21.8)	62.75	20.41
Whether they had visited the emergency room, n (%)				0.58	0.57
Yes	72 (21.8)	65.83	20.35
No	258 (78.2)	64.27	19.87
Children’ month of age, mean±SD	15.07±8.18			−0.06	0.26
The number of identified food allergens, median (IQR)			2 (2)	0.197	<0.001***
The number of highly suspected food allergens, median (IQR)			2 (2)	0.219	<0.001***
The number of children’s FA-affected systems,d median (IQR)）			2 (1)	0.213	<0.001***
The number of children’s FA-related symptoms, median (IQR)			2 (3)	0.252	<0.001***

* p<0.05; **p<0.01; ***p<0.001.

aChildren did not take in special formula milk powder, only took in breast milk or formula milk.

bSevere anaphylaxis refers to acute systemic allergic reactions that can be life-threatening, including anaphylactic shock, angio-oedema, bronchospasm and hypotension.

cThe antiallergy drugs referred to here are antihistamines. In China, patients are not prescribed self-injectable epinephrine.

dFA-affected systems involve the integumentary system, respiratory system and gastrointestinal system.

FA, food allergies.

Table 2

General characteristics of parents having children with food allergies and the relationships between the characteristics and parental burden (n=330)

Characteristics		Parental burden
		Mean	SD	t/F/r	P value
Gender, n (%)				−0.37	0.71
Male	19 (5.8)	62.95	22.29
Female	311 (94.2)	64.71	19.84
Parental history of self-reported allergies, n (%)				−0.08	0.94
Yes	211 (67.0)	64.55	19.29
No	109 (33.0)	64.73	21.34
Education level, n (%)				5.79	<0.01**
Junior high school or below	12 (3.6)	82.75	16.48
Senior high school	28 (8.5)	72.68	18.15
College	95 (28.8)	67.02	20.44
Undergraduate	158 (47.9)	61.44	19.73
Graduate or above	37 (11.2)	59.95	16.68
Working conditions, n (%)				−4.03	<0.01**
At work	231 (70.0)	61.77	20.44
Not at work	99 (30.0)	71.22	17.12
Financial affordability regarding their children’s FA, n (%)				15.04	<0.01**
Fully affordable	62 (18.8)	56.69	20.14
Able to afford	195 (59.1)	62.56	19
Harder to afford	67 (20.3)	76.1	17.22
Unable to afford	6 (1.8)	84.5	12.11
Knowledge about FA (self-reported), n (%)				−1.36	0.17
Yes	113 (34.2)	62.54	18.31
No	217 (65.8)	65.69	20.72
Whether they were obtaining knowledge about FA from health professionals, n (%)				−1.85	0.07
Yes	233 (70.6)	63.3	20.22
No	97 (29.4)	67.75	19.04
Whether they had a medical background, n (%)				−0.77	0.45
Yes	56 (17.0)	62.75	20.14
No	274 (83.0)	64.99	19.93
Parents’ age (years), mean±SD	30.41±3.68			−0.06	0.26

*p<0.05.**p<0.01.

FA, food allergies.

Analysis of factors influencing parental burden among parents of children with FA

Univariate analysis on factors related to parental burden among parents of children with FA

The results in table 1 show that parents whose children were allergic to eggs had a higher parental burden score (67.42±17.57) than those whose children did not (62.15±21.28), and parents whose children had a history of antiallergy drug use (67.14±19.10) had a higher parental burden score than those whose children did not (62.75±20.41). In addition, the parental burden increased with the number of children’s identified food allergens, the number of highly suspected food allergens, the number of children’s FA-affected systems and the number of children’s FA-related symptoms. The intake of children’s special formula milk affected the parental burden score, and parents whose children took in amino acid milk powder had the highest parental burden (70.67±17.97). Moreover, parents whose children had up to 10 times adverse reactions (70.18±19.80) had higher parental burden scores than parents whose children had 1–5 times (62.77±20.33, p<0.01) and 6–10 times adverse reactions (60.07±17.63, p<0.01). As table 2 shows, the parental burden scores among the parents of children with FA decreased with educational level and financial affordability regarding their children’s FA. Moreover, parents who were not at work (71.22±17.12) had a higher parental burden than those who were at work (61.77±20.44). In addition, all of them reported significant statistical differences (p<0.05). However, no significant difference in parental burden existed between groups divided by children’s and parents’ genders, when children had visited the emergency room, when children had experienced severe anaphylaxis, when there was a parental history of self-reported allergies, etc. The bivariate correlation analysis results in table 3 show that parental burden was negatively correlated with the parents’ social support (r=−0.291, p<0.001) and coping style (r=−0.240, p<0.001). Because the analysis above did not control other confounding factors, these statistically significant factors in this part need to be identified in multiple stepwise linear regression.

Table 3

Bivariate correlation between parental burden and social support and coping style (n=330)

	Parental burden
	Spearman correction coefficients (r)
Social support	−0.291***
Coping stylea	−0.240***

***p<0.001.

aIt represents the coping style tendency (identified after Z-conversion), which means the individual’s main coping style in stress states.

FA, food allergies.

Multiple linear regression analysis on factors related to parental burden among parents of children with FA

Some factors were introduced in a multiple stepwise linear regression to ensure a stable model. The children-related factors included the number of identified food allergens, the number of highly suspected food allergens, the number of children’s FA-affected systems, the number of children’s FA-related symptoms, the frequency of adverse reactions, whether they had an intake of special formula milk powder, whether they had an egg allergy, whether they had experienced severe anaphylaxis and whether they had a history of antiallergy drug use. Parent-related factors included education level, working conditions, financial affordability regarding their children’s FA, knowledge of FA (self-reported), medical knowledge background, social support, coping style and whether they were obtaining learning knowledge from health professionals. ‘Intake of special formula milk powder-None’ was set as a reference variable for ‘Intake of special formula milk’ in introducing dummy variables into regression analysis. In table 4, the results show that the number of highly suspected food allergens (β=0.162, p<0.01), the number of children’s FA-affected systems (β=0.129, p<0.05), amino acid milk powder intake (β=0.154, p<0.01), working conditions (β=−0.169, p=0.000), financial affordability regarding their children’s FA (β=−0.224, p=0.000) and social support (β=−0.226, p=0.000) are independently associated with the parental burden on parents of children with FA. Coping style was not entered into the regression equation in this study. The multiple linear regression equation models were statistically significant (p<0.001, F=6.769) and explained 27.4% of the variance (adjusted R2 was 26.0%).

Table 4

A multivariate linear stepwise regression analysis of the factors influencing the parental burden of parents having children with food allergies (n=330)

Coefficient
Model	Unstandardised coefficient		Standard coefficient	t	P value
	B	SE	Beta
(Constant)	61.438	7.225		8.47	0.000***
The number of highly suspected food allergensa	13.489	4.127	0.162	3.269	0.001**
The number of children’s FA-affected systemsa	20.118	7.733	0.129	2.602	0.010*
Intake of special formula milk powder—amino acid milk powder intake	7.328	2.292	0.154	3.198	0.002**
Whether they had an intake of special formula milk powder—noneb	0
Working conditions	−7.346	2.082	−0.169	−3.529	0.000***
Financial affordability regarding their children’s FA	−6.568	1.454	−0.224	−4.517	0.000***
Social support	−0.58	0.127	−0.226	−4.582	0.000***

R2=0.274; adjusted R2=0.260; △F=7.769; *p<0.05; **p<0.01; ***p<0.001; dependent variable: parental burden among parents with an FA child. The variables with p<0.1 from univariate analysis were introduced.

aThe variables introduced in a multiple stepwise linear regression here were log-transformed (y=log10 (x+1)) to approximate a normal distribution.

bIt was set as a reference variable for ‘Intake of special formula milk’ in the process of introducing dummy variables into regression analysis.

FA, food allergies.

Discussion

Providing quality care for children with FA will undoubtedly place a burden on their parents. The purpose of the study was to explore the associated factors of parental burden among the parents of children with FA. In our sample, the score for parental burden was 64.61 (SD=19.96), which was lower than the reported in the UK (69.99 (SD=23.17).47

Among children-related characteristics, it was determined that the number of highly suspected food allergens and the number of FA-affected systems positively influenced the parental burden among parents of children with FA. These results are consistent with previous studies.17 30 47 When children have more highly suspected food allergens, their parents need to spend more effort to take a wide range of avoidance measures and often feel worried, which increases the parental burden. In addition, when FA affects many systems involving the skin and the gastrointestinal, respiratory and cardiovascular systems, parents feel more worried about their children’s health.11 In one of the noteworthy findings of this study, the parental burden of parents whose children took in amino acid milk powder was high. This may be related to the fact that parents were worried about the lack of pure nutrient intake when amino acid milk powder was an alternative to breast milk and cow’s milk.48 There is a great need for comprehensive and systematic FA-based health education programmes.49–51 On the one hand, it will help parents manage their children’s food allergies (FA) correctly and find alternative nutrients to ensure normal growth and development.52 53 On the other hand, it will change wrong cognitions related to FA to decrease the parental burden among parents of children with FA.50 Additionally, amino acid milk powder is more expensive than ordinary powdered milk. Because FA-related economic aid is not provided in China, parents whose children consumed amino acid milk powder must spend more to provide it, which would increase their financial burden.

Among the parent-related characteristics, this study revealed that the parental burden of parents who were not at work was high, which coincides with the results in the previous literature.54 55 This situation can be interpreted in such a way that parents at work have higher economic power, a better social environment and better use of their social support systems.56–58 When parents do not work and devote themselves to the care of the FA child, they may lack social support and develop psychosocial symptoms, which may increase the burden.11 13 This study also showed that the lower the financial affordability regarding their children’s FA, the heavier the parental burden. This was consistent with the results of a study focusing on the economic impact of childhood FA in the USA.58 The high costs, such as direct medical costs, dietary costs and opportunity costs (such as restricted career choices), definitely increase the burden on parents and may not be affordable for some families. Couch et al suggested that earlier OFCs should be considered to identify food allergens and reduce families’ additional financial expenses, especially for opportunity costs. This would effectively decrease the financial burden on parents.59 Furthermore, there is an urgent need for providing a robust health insurance system for relieving their parental burden financially.60 The present study found that the gender of parents was not an influencing factor associated with parental burden, which differs from the results of Springston et al.18–20 This may be because the sample size of the fathers in this study was small, which may have led to the difference in parental burden between the mothers and fathers not being identified. The reason for the small sample size of fathers recruited in this study is also related to the societal context of China, where women often assume the role of childcare.61 62

With the highest standardised beta in the regression equation, parents’ social support was a key protective influencing factor against parental burden among parents of children with FA. Specifically, parental burden decreases as parents’ social support increases. Previous studies have also suggested that when parents (children’s caregivers) lack social support during the caregiving process, they may develop physical symptoms, such as weight loss and fatigue, and psychosocial symptoms, such as anxiety, social isolation and decreased quality of life, which all increase the burden of care.63–65 Adequate social support may help reduce the parental burden. Social support has traditionally been provided by families and friends in China and less provided by the community service agency and the government.37 66 67 In a qualitative study concerning caregivers caring for a family member, 62% of respondents identified a lack of social support as one of their caregiving challenges. Moreover, 83% of participants identified emotional support (ie, seeking out to friends and family or others) to be effective in dealing with these challenges.68 This individual level of social support will help parents establish and improve their social support system so that they can feel more supported while dealing with the problems and emotional reactions related to their children’s FA.64 69 Parents can also use social media platforms and websites to share information and learn more about FA.70 71 In addition, community support services, such as health professional help, structured educational programmes and workshops related to the management of food allergens, should be available to parents of children with FA.72 They will provide better community social support and richer knowledge about managing FA for parents and help them cope with caring for a child with FA, thus relieving the parental burden.50 73 Additionally, health-related departments should improve medical policies and reimbursement rates to provide financial support for parents with children with FA.60

However, parents’ coping style did not exhibit a relationship with parental burden in the multiple stepwise regression of this study. The reason may be that although some parents adopted an active coping style (eg, positively learning knowledge about their children’s FA and avoiding allergens as much as possible) to deal with their children’s FA, their specific, positive measures cannot immediately cure or control their children’s allergy reactions.70 Furthermore, allergic reactions are unpredictable and, for many children, FA is sometimes accompanied with food intolerance for which the symptoms are similar to those of the FA and difficult to avoid.74–77 In these conditions, an active coping style cannot always obtain positive feedback. This would cause anxiety and a heavy burden, especially an emotional burden, on parents.14 70 In addition, the SCSQ used to measure coping style focused on only two types. However, coping styles are much more complex than merely positive and negative. The quantitative measurement instrument we used ignored the complexity of coping styles, which may have influenced the final results. Therefore, instruments that can identify the complexity of coping styles should be developed. Furthermore, qualitative studies should be considered to explore the relationship between coping style and parental burden.

Limitations

Several limitations of this study are acknowledged. Considering we only recruited the parents with a child aged 0–5 years with FA, the generalisation of the results in this study requires caution for other populations. Second, convenience sampling cannot guarantee sample diversity. For example, the sample size of fathers is limited in this study. However, it is also consistent with the fact that the mother is primarily responsible for the care of the sick child in China. To better explore the fathers’ parental burden, further studies need recruit more fathers in the study sample. Third, it was certainly a limitation that FA diagnosis was made using diverse methods and not always based on the gold standard of OFC. In middle-income and low-income countries such as China, FA diagnosis is likely to be based on the skin prick test, serum IgE detection or clinical history, while OFC is rarely used for diagnosis. Sometimes misdiagnosis may occur. Finally, the quantitative measurement instrument of coping style employed in this study ignores the complexity of coping styles, which may have influenced the results. Instruments that can identify the complexity of coping styles should be developed and used in future studies.

Conclusion

Parents whose children had more highly suspected food allergens and had more FA-affected systems had a higher parental burden. Moreover, the burden of parents whose children took in amino acid milk powder was higher than that of parents whose children did not. Parents’ working conditions were also a significant influencing factor of parental burden. All these factors could greatly predict parental burden among parents with children with FA. Above all, parents with lower financial affordability regarding their children’s FA and lower social support experienced higher parental burdens. Thus, the level of financial affordability regarding their children’s FA and parents’ state of social support can serve as key predictors of parental burden among parents of children with FA. These findings point to several measures that can help ease the parental burden. FA should be included within the scope of medical insurance reimbursement, which can partially alleviate financial pressure on parents. Additionally, high-quality FA-related health education programmes should be provided by community health service centres to support parents in monitoring and flexibly responding to children’s FA. Moreover, it can reduce the daily life and social burden of parents with children with FA, thereby promoting their physical and mental health. This study provided basic evidence to explore strategies for relieving the parental burden among parents of children with FA. In the future, more empirical studies (eg, longitudinal studies and intervention studies) are needed to verify and further the results of this study.